Data Exploration - Calories Project

library(tidyverse)
library(dplyr)
# install.packages("skimr")
library(skimr)
# install.packages("janitor")
library(janitor)
# install.packages("ggcorrplot")
library(ggcorrplot)

# load clean file
life_clean <- read.csv("../data/clean/life_clean.csv")

Error in file(file, "rt") : cannot open the connection

# looking at calories_burned as the y variable
# calories_burned: total calories burned during the session

summary(life_clean$calories_burned)

   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
  323.1   910.8  1231.5  1280.1  1553.1  2890.8 

ggplot(life_clean, aes(x = calories_burned)) +
  geom_histogram(bins = 30, color = "purple", fill = "lightblue4")

# looking at a right skewed distribution with a mean about 1280 total calories burned

ggplot(life_clean, aes(x = calories_burned)) +
  geom_boxplot(fill = "steelblue")

# boxplot shows some outliers on the higher calorie burn end

# looking at burns_calories_per_30min as the y variable
# burns_calories_per_30min: estimated calories burned in 30 min of "that" exercise

summary(life_clean$burns_calories_per_30min)

   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
  128.8   337.6   348.6   344.0   360.5   381.7 

ggplot(life_clean, aes(x = burns_calories_per_30min)) +
  geom_histogram(bins = 30, color = "purple", fill = "lightblue4")

# looking at a left skewed distribution with a mean about 344 calories burned in a 30min session

ggplot(life_clean, aes(x = burns_calories_per_30min)) +
  geom_boxplot(fill = "steelblue")

# boxplot shows an even distribution of outliers on the lower calorie burn 

Explore Diet Variables

Using calories_burned for the dependent variable

ggplot(life_clean, aes(x = carbs, y = calories_burned)) +
  geom_point() +
  geom_smooth(method = "lm")

ggplot(life_clean, aes(x = proteins, y = calories_burned)) +
  geom_point() +
  geom_smooth(method = "lm")

ggplot(life_clean, aes(x = fats, y = calories_burned)) +
  geom_point() +
  geom_smooth(method = "lm")

ggplot(life_clean, aes(x = calories, y = calories_burned)) +
  geom_point() +
  geom_smooth(method = "lm")

ggplot(life_clean, aes(x = sugar_g, y = calories_burned)) +
  geom_point() +
  geom_smooth(method = "lm")

ggplot(life_clean, aes(x = diet_type, y = calories_burned)) +
  geom_boxplot()

ggplot(life_clean, aes(x = diet_type, y = calories_burned)) +
  geom_point() +
  geom_smooth(method = "lm")

ggplot(life_clean, aes(x = burns_calories_bin, y = calories_burned)) +
  geom_point() +
  geom_smooth(method = "lm")

mean(life_clean$carbs)

[1] 249.7786

mean(life_clean$proteins)

[1] 99.91829

mean(life_clean$fats)

[1] 66.61217

mean(life_clean$calories)

[1] 2024.418

mean(life_clean$sugar_g)

[1] 24.53104

mean(life_clean$cholesterol_mg)

[1] 148.4656

mean(life_clean$cal_balance)

[1] 744.3087

Explore Exercise Variables

Using calories_burned for the dependent variable

ggplot(life_clean, aes(x = max_bpm, y = calories_burned)) +
  geom_point() +
  geom_smooth(method = "lm")

ggplot(life_clean, aes(x = resting_bpm, y = calories_burned)) +
  geom_point() +
  geom_smooth(method = "lm")

ggplot(life_clean, aes(x = session_duration_hours, y = calories_burned)) +
  geom_point() +
  geom_smooth(method = "lm")

ggplot(life_clean, aes(x = workout_type, y = calories_burned)) +
  geom_point() +
  geom_smooth(method = "lm")

ggplot(life_clean, aes(x = workout_frequency_days_week, y = calories_burned)) +
  geom_point() +
  geom_smooth(method = "lm")

ggplot(life_clean, aes(x = sets, y = calories_burned)) +
  geom_point() +
  geom_smooth(method = "lm")

ggplot(life_clean, aes(x = reps, y = calories_burned)) +
  geom_point() +
  geom_smooth(method = "lm")

ggplot(life_clean, aes(x = workout_type, y = calories_burned)) +
  geom_boxplot()

num_vars <- life_clean |> select(where(is.numeric))

corr <- cor(num_vars, use = "complete.obs")

ggcorrplot::ggcorrplot(corr, lab = FALSE, hc.order = TRUE, tl.cex = 6, tl.srt = 45)